Retro-Technics • author: a3310i • last modified: 2022.02.22 •

There have been huge advances in electronics over the past 30 years. Professor Stuart Madnick of the Massachusetts Institute of Technology has stated that if similar advances were made in the field of automobiles, a Rolls Royce car would use a gallon of gasoline (3.8 litres) for 2 million kilometres and would cost $2.5! This staggering comparison gives an excellent idea of the dynamics of the processes taking place in the development of electron technology. The level of technology, especially electron technology, is characterised by three basic parameters:

The value of each of these parameters has improved many times over the decades and, as a result of the process of perfecting electron technology, the conditions have arisen to produce a microprocessor circuit.
The history of computing is as old as human history itself. However, the history of the automation of computing only began in the 17^th century, when industry and commerce began to develop on an unprecedented scale. It was then that the first mechanical calculating machines were constructed, the most famous of which is the calculator of the famous French mathematician, physicist and philosopher Blaise Pascal. Various types of mechanical calculators greatly facilitated people's computing work over the next three centuries, especially in the 19^th century, when the production of mechanical calculators was developed on an industrial scale, so that they found their way into the coffers of every industrial, commercial or banking company. The concept of automation of calculations was then greatly developed by the English mathematician Charles Babbage, who devoted practically his entire life to the development of two designs of mechanical calculating machines and the methodology of carrying out calculations with their help. The structure of the second of these machines, the so-called Analytical Machine, bears a striking resemblance to a modern computer and, in retrospect, Babbage's concept should be regarded as epoch-making. Unfortunately, Babbage's ideas were not realised for financial and technological reasons and were forgotten. The first half of the 20^th century was marked, on the one hand, by the rapid development of electrical and electron technology and, on the other, by the creation of theoretical foundations in technology and science that would, in future, make possible the construction and programming of computers.

[001] Electromagnetic relay design diagram: 1 - electromagnet, 2 - anchor, 3 - contact, 4 - contact, I₁ - input current, I₂ - output current

[002] Characteristics of the electromagnetic relay: I₁ - input current, I₂ - output current, I_ts - treshold current

The three basic parameters characterising the switching element for an electromagnetic relay were as follows: volume 85-180 cm³, induction power 3-20 mW, frequency less than 60 switches per second, which corresponds to a delay time of about 20 ms. At the turn of the thirties and forties elctromagnetic relays with the above-mentioned parameters were used for the construction of electric calculators. The need to build fast calculating devices resulted from the turbulent development of the American economy and military needs. In 1937 two teams independently of each other started to work on relay calculators. The first of these, operating within the Bell Telephone company under the leadership of mathematician Georg R. Stibiz, built a calculator known as the Model I in 1939. The second, selected from among IBM specialists and led by Howard Aiken, realised Charles Babbage's concept of a mechanical calculator in an electric version. This calculator, known as the Mark 1, was ready in 1944. The size of this machine was impressive. It was 17 m long and about 3 m high, consisting of 800,000 parts. The Mark I operated on twenty-three-digit figures, adding them in 0.3 seconds and multiplying them in 3 seconds. The machine was able to calculate the values of elementary logarithmic and trigonometric functions. However, the speed of electromechanical calculators was not sufficient to solve many complex technical, scientific, economic and especially military problems. The outbreak of World War II significantly intensified research work on automated calculating devices. The development of rocket and nuclear technology required devices at least twice as fast as the Mark 1 calculator.

[003] Vacuum tube wiring diagram.

The first construction of an electron tube that could act as a switching element, the so-called triode, was made in 1906 by Lee de Forest. The circuit diagram of a triode tube is illustrated in Figure 003. The principle of digital triode use is very simple. As long as a sufficiently large potential of negative sign is applied to the central electrode and the so-called grid, the tube remains in a state of i.i.d. occlusion. A step change of the grid potential in the positive direction causes a large number of electrons to flow from the cathode to the anode and a step change in the anode current is produced. At this point the lamp is in the conduction state. The possibility of a step change of state led to the use of electron tubes in the construction of counting devices. In 1943, two scientists from the University of Pennsylvania in Philadelphia, J. Prosper Eckert and John W. Mouchly, proposed the construction of a calculating machine using new switching elements - electron tubes. As a result of their work, a calculating device, first called a computer, was created in early 1946. This was the Electronic Numerical Integrator and Computer - known as ENIAC. The machine consisted of 18,000 tubes and 1,500 relays, occupying a 9x15 m room, weighed 30 tons and consumed 150 KW of electrical power. It was indeed 1000 times faster than the best relay calculators.
Moreover, although the dimensions of the lamps were not much different from those of the relays, their weight was seriously reduced. The use of the new switching element not only increased the speed brilliantly, but also the reliability of the whole device. This was due to the absence in the electron tube of any moving mechanical element such as the anchor and contacts in the relay.

[004] First generation computer built on electron tubes.

A breakthrough in the development of many fields of technology, and especially in the construction of computers, was the discovery in 1947 by three American scientists: William Shockley, John Bardeen and Walter H. Brattain of a new semiconductor electronic element - the transistor. Since then, the continuous progress of semiconductor technology has led to the conditions for the construction of the first microprocessor. All the parameters of the new switching element proved to be better than those of the tubes, and in particular the dimensions and the power required for its operation were considerably miniaturised. The first transistors were made in bipolar technology. This meant that two carriers participated in the conduction of current: electrons and holes, which represent the state of no electron in a semiconductor material and have a positive charge. Further significant miniaturization of the transistor switching element occurred with the development of unipolar technologies where current is created by only one type of carriers: electrons or holes. Unipolar transistor technology is based on a combination of metals, oxides and semiconductors and is called MOS technology (Metal-Oxide-Semiconductor). Depending on the carrier type, it is described by a specific prefix: n if it is an electron, p if it is a hole.
A logical consequence of the development of unipolar technologies was the idea of placing a larger number of cooperating switching elements on a single semiconductor crystal. In 1901 the Fairchild Semiconductors company produced on an industrial scale the first integrated circuit in planar technology, which contained a flip-flop composed of four transistors, and thus a circuit capable of storing an elementary particle of information, the so-called bit (it takes the value of 0 or 1). The sixties resulted in a large number of integrated circuit designs containing from a few to several hundred elements on a single structure. An outstanding example of a whole series of such circuits is the series produced by Texas Instruments, called SN 74.
Integrated circuits, including microprocessors, are currently manufactured using a variety of technologies that differ in their basic parameters. Users of these chips most often demand that the following requirements are met:

- high speed of operation
- low power consumption
- high component density

These requirements are usually in conflict with each other and different technologies are better or worse due to certain parameters of which the basic ones are:

t - delay of a signal caused by its passage through a switching element or a group of them realizing an elementary logic function (gate)
P - power consumption by the elementary gate
P*t - product of the energy required to operate the elementary gate
K - packing density of switching elements

Circuits constructed using the most advanced technologies have the capacity to perform millions of complex operations per second. However, currently solving most technical or scientific problems requires building computers performing many billions of operations per second. The search is therefore on for ever faster switching elements within the semiconductor technologies themselves. Here, high hopes are pinned on gallium arsenide, which makes it possible to construct elements that delay signals by as little as 100 picoseconds (100*10^-12 s). Within other technologies, high hopes are placed on Josephson junctions, made using superconductor technology, which delay signals by around 10 ps, and light transmission elements that delay signals by single picoseconds.